Hemodialysis vascular access dysfunction accounts for extensive morbidity, and is the greatest contributor to cost among patients with renal failure. Diabetics and African Americans (AA's) are disproportionately affected;however, reasons for this are unclear. The primary aims of the proposed study are 1) estimate the association between diabetes and AA race and arteriovenous fistula (AVF) survival;2) explore whether factors associated with vessel location and primary cause of renal disease account for the relationship between diabetes and AA race and AVF survival;3) determine if vein caliber and arterial resistance account for the association between diabetes and AA race and AVF survival, and 4) determine if inflammation accounts for the association between diabetes and AA race and AVF failure. Two parallel activities will be conducted to address our scientific aims. First we will use data from the Centers for Medicare and Medicaid end-stage renal disease (ESRD) Clinical Performance Measures (CPM) Project, United States Renal Data System (USRDS) and an established cohort from a Veterans'Affairs sponsored study of dialysis patients undergoing AVF placement to identify factors that might account for the association between diabetes and race and AVF survival. Second, we will recruit a cohort of 180 pre-dialysis patients preparing to undergo access placement from the clinics at Emory University, Emory Crawford Long Hospital, Grady Memorial Hospital and the Dialysis Access Center of Atlanta to ascertain if serum markers of inflammation account for the association between diabetes and AA race and access survival. The proposed studies aim to elucidate reasons for the disparities in vascular access survival among diabetics and AA's.